CONCLUSION

The pair trading models employing simple linear regression and ARIMA methods for calculating control limits offer distinct advantages and face specific limitations. The simplicity and interpretability of the simple linear regression model make it easy to implement, providing a straightforward relationship between the two stocks. However, it assumes a linear relationship and is sensitive to outliers, potentially impacting accuracy. In contrast, the ARIMA model is tailored for time series data, capturing temporal patterns and offering statistical rigor for model validation. Yet, ARIMA models are more complex, require stationary data, and are better suited for short to medium-term predictions. The effectiveness of each model depends on the stability of the correlation between stock prices over time, with constant reassessment and adaptation to changing market conditions essential. Thorough model validation, backtesting, and the incorporation of risk management strategies are crucial for robust pair trading strategies. Additionally, considering factors beyond stock prices, such as market indices or economic indicators, can enhance the models' comprehensiveness. Overall, the choice between these models hinges on the specific characteristics of the data and the desired level of complexity, with continuous monitoring and refinement being imperative for successful pair trading strategies.